© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 142638
    Prev
    Next
    INDEX
    Explanations

    Long, diverse texts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ç¯ĩ
    -0.30
    åŃ£
    -0.28
    xAF
    -0.26
    atern
    -0.25
    emens
    -0.25
    etas
    -0.25
    ascript
    -0.25
    iem
    -0.24
    责任
    -0.24
    æĬĢèĥ½åŁ¹è®Ń
    -0.24
    POSITIVE LOGITS
    ï¾Ł
    0.28
    .setLayout
    0.28
    ä¸įæĿ¥
    0.27
     augmentation
    0.26
    SSERT
    0.26
    ologies
    0.25
    害
    0.24
    èŁł
    0.24
    à¹ĥà¸Ī
    0.24
    deb
    0.23
    Activations Density 42.664%

    No Known Activations